Baldur: Whole-proof generation and repair with large language models

E First, MN Rabe, T Ringer, Y Brun - Proceedings of the 31st ACM Joint …, 2023 - dl.acm.org
Formally verifying software is a highly desirable but labor-intensive task. Recent work has
developed methods to automate formal verification using proof assistants, such as Coq and …

Diversity-driven automated formal verification

E First, Y Brun - Proceedings of the 44th International Conference on …, 2022 - dl.acm.org
Formally verified correctness is one of the most desirable properties of software systems. But
despite great progress made via interactive theorem provers, such as Coq, writing proof …

Can Transformers Learn to Solve Problems Recursively?

SD Zhang, C Tigges, S Biderman, M Raginsky… - arXiv preprint arXiv …, 2023 - arxiv.org
Neural networks have in recent years shown promise for helping software engineers write
programs and even formally verify them. While semantic information plays a crucial part in …

Proof repair infrastructure for supervised models: Building a large proof repair dataset

T Reichel, R Henderson, A Touchet… - … Proving (ITP 2023), 2023 - drops.dagstuhl.de
We report on our efforts building a new, large proof-repair dataset and benchmark suite for
the Coq proof assistant. The dataset is made up of Git commits from open-source projects …

SwitchV: automated SDN switch validation with P4 models

KD Albab, J DiLorenzo, S Heule… - Proceedings of the …, 2022 - dl.acm.org
Increasing demand on computer networks continuously pushes manufacturers to
incorporate novel features and capabilities into their switches at an ever-accelerating pace …

Transformer-based models are not yet perfect at learning to emulate structural recursion

D Zhang, C Tigges, Z Zhang, S Biderman… - arXiv preprint arXiv …, 2024 - arxiv.org
This paper investigates the ability of transformer-based models to learn structural recursion
from examples. Recursion is a universal concept in both natural and formal languages …

Mostly Automated Proof Repair for Verified Libraries

K Gopinathan, M Keoliya, I Sergey - Proceedings of the ACM on …, 2023 - dl.acm.org
The cost of maintaining formally specified and verified software is widely considered
prohibitively high due to the need to constantly keep code and the proofs of its correctness in …

Proofster: Automated formal verification

A Agrawal, E First, Z Kaufman, T Reichel… - 2023 IEEE/ACM 45th …, 2023 - ieeexplore.ieee.org
Formal verification is an effective but extremely work-intensive method of improving software
quality. Verifying the correctness of software systems often requires significantly more effort …

Fermat's Last Theorem for regular primes

AJ Best, C Birkbeck, R Brasca, ER Boidi - arXiv preprint arXiv:2305.08955, 2023 - arxiv.org
We formalise the proof of the first case of Fermat's Last Theorem for regular primes using
the\emph {Lean} theorem prover and its mathematical library\emph {mathlib}. This is an …

Pip-MPU: Formal verification of an MPU-based separation kernel for constrained devices

N Dejon, C Gaber, G Grimaud - International Journal of Embedded …, 2023 - hal.science
Pip-MPU is a minimalist separation kernel for constrained devices (scarce memory and
power resources). In this work, we demonstrate high-assurance of Pip-MPU's isolation …